No abstract

The popular adage “publish or perish” has long defined individual career strategies as well as scholarly investigations of earnings inequality in academe, as researchers have relied heavily on research productivity to explain earnings inequality among faculty members. Academia, however, has changed dramatically in the last few decades: it has become larger and more demographically diverse, and fears of overspecialization prompt calls for interdisciplinary approaches. In this new environment, other factors, in addition to productivity, are likely relevant to our understanding of earnings differentials. In this article, I assess whether two additional factors—visibility and the extent of research specialization—contribute to men's earning advantage. Using probability samples of tenure-track academics in two disciplines, a variety of data sources, and innovative measures, I find that both factors are highly relevant to the process by which earnings are determined. Women earn less than men largely because they specialize less. Lower levels of specialization hinder productivity, productivity enhances visibility, and visibility has a direct, positive, and significant effect on salary. I discuss the practical implications of these findings and lay the foundation for a broader theory of the role of research specialization in work processes.

This article reviews trends in the practice and study of research collaboration, focusing on journal publications in academic science. I briefly describe the different styles and types of collaboration and then focus on the drivers of the trend toward increased collaboration and on its consequences for both individual researchers and science more generally. Scholarship on collaboration seems partial to delineating its benefits; this review highlights the increasing body of research that focuses instead on the possible costs of collaboration. The synthesis reveals several topics that are ripe for investigation, including the impact of collaboration on the contributing authors and their work, the use of multiple methods and measures, and research integrity. I applaud a few recent efforts to overcome the perennial file-drawer problem by gaining access to collaborations that do not result in publication and thus are typically removed from public review and the research analyst's eye.

To date, studies on how having children affects the research productivity of academics, and whether the effects differ by gender, have had inconsistent findings. Using nuanced measures of parental obligations and linear growth modeling, we analyzed the effects of children on the entire careers of academics in two disciplines — linguistics and sociology — and tested for differential effects by gender. In addition, we modeled not only productivity, but also visibility, another component of scholarly success. Our findings suggest that after the birth a child, productivity growth declines, but more so for women. Thus, children account for part of the gender gap in rates of productivity over time. Children also have an impact on the research visibility of academics, but cannot account for gender differences in visibility.

To characterize the nature of innovation, we draw on foundational theories of scientific and technological change 2,29,30 , which distinguish between two types of breakthroughs. First, some contributions improve existing streams of knowledge, and therefore consolidate the status